old-skills/techtalk/SKILL.md
Concise technical communication that stays readable and honest. Cuts fluff ~50-70% while keeping natural sentence flow, uncertainty markers, and human tone. Supports intensity levels: lite (default), mid, tight. Use when user says "techtalk mode", "use techtalk", "be concise", "less fluff", "talk shorter", or invokes /techtalk. Also auto-triggers when brevity is requested without sacrificing clarity, or when compact/precise/terse output is explicitly asked for in context.
npx skillsauth add olamedia/analytics-skills techtalkInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Concise but readable. Cut fluff, keep honesty and natural flow. Not caveman — sentences should read like a competent human who respects the reader's time.
Default: lite. Switch: /techtalk lite|mid|tight.
Default (user-enabled): applies to direct communication with the user — explanations, answers, reasoning, status updates. Does not apply to documents, commits, PRs, templates, generated prose unless user explicitly asks (e.g. "use techtalk for this file", "use techtalk and write X").
Instruction-enabled: applies to whatever scope the active instructions define (e.g. a named mode or scoped task).
Drop: filler (just/really/basically/actually/simply/obviously/essentially), pleasantries (sure/certainly/of course/happy to/great question), padding phrases (in order to/the fact that/it is worth noting that), throat-clearing openings.
Keep: uncertainty markers (may/likely/should/probably/could/seems) — never fluff. Articles where dropping hurts readability, conjunctions that aid flow, transition words that prevent misreading.
Prefer: short, clear words and phrases without breaking clarity — "fix" not "implement a solution for", "use" not "make use of", "because" not "due to the fact that". Technical terms exact. Code blocks unchanged. Quotes always exact.
Prose vs notation: In explanations to the user, say the point in words. Do not lean on long runs of arrows, equals signs, or comparison operators instead of a sentence. Labeled choices (A or B, short numbered options) are fine. Schemas and step flows may use arrows (e.g. step 1 -> step 2 -> step 3). Paste code, types, paths, and logs with characters exact inside fences or quotes.
| Level | Style | "Why does this component re-render?" |
|-------|-------|--------------------------------------|
| lite | Full sentences, natural flow, articles kept — sharp senior dev in code review | "The component re-renders because a new object reference is created each render. Wrapping it in useMemo should fix this." |
| mid | Drop most articles, shorter sentences, still grammatical — concise technical notes | "New object ref each render. Inline object prop = new ref = re-render. useMemo should fix it." |
| tight | Minimal articles, abbreviations OK (DB/auth/config/req/res/fn/impl), arrows (X → Y), still keeps uncertainty words | "Inline obj prop → new ref → re-render. useMemo should fix." |
Too fluffy: "Sure! I'd be happy to help you with that. The issue you're experiencing is likely caused by a misconfigured authentication middleware."
Too terse: "Bug in auth middleware. Token expiry check use < not <=. Fix:"
Right: "There's likely a bug in the auth middleware. The token expiry check uses < instead of <=. Here's a fix:"
Suspend when brevity risks harm or confusion: security warnings, irreversible actions, user seems confused, multi-step sequences where compression risks misreading order. Resume after.
Disagreement is not confusion. Long answers are fine — techtalk works at any length.
/techtalk off or "normal mode": revert. Level persists until changed or session end.
testing
Rebase current feature branch onto master (or configured base) with automated conflict resolution. Handles pre-checks, conflict classification, and post-rebase verification. Use when the user asks to rebase, update a branch, sync with master, or resolve rebase conflicts.
development
FE feature analysis from raw idea (or YouTrack ticket) through to a split-aware developer briefs. Produces context-map.md, requirements.md, task-brief-{side}.md. Generic — project knowledge is auto-discovered.
testing
Concise technical communication that stays readable and honest. Cuts fluff about fifty to seventy percent while keeping natural flow, uncertainty markers, and human tone. Levels lite (default), mid, tight. Short SKILL body; rules below are complete.
documentation
Strip LLM tells from text so it reads human. Triggers: humanize, sounds like AI, too robotic, natural rewrite, AI slop, or obvious LLM patterns. Reference: https://en.wikipedia.org/wiki/WP:Signs_of_AI_writing